[gmx-users] density profile

Justin Lemkul jalemkul at vt.edu
Sun Mar 31 19:23:10 CEST 2013



On 3/31/13 1:12 PM, Elisabeth wrote:
> Thanks Justin for your reply. To avoid compressing down the cell I thought
> using semiisotropic option with 0 compressibility in Z would be
> appropriate.
>
> Here are the steps:
>
> 1- First I did a 10 ns NPT to equilibrate the box and used the last frame
> gro and cpt file as input for the next step
> 2- I extended the box in Z to more than twice the initial box size and
> issued the following:
>
> grompp -f md.mdp -c extendedZ.gro -t .cpt -p .top -o extendedZ.tpr
> mpirun -np 4 mdrun_mpi -deffnm .extendedZ -s -o -c -g -e -x -v
>
> My questions are:
>
> 1) Can I get the proper solvent-air interface to obtain density profile by
> extending the Z direction (last line of the gro file obtained from NPT at
> each pressure)? Is that what yo mean by " then build a new unit cell and
> run under NVT" ?

That's one approach.  It's not NVT if you're using pressure coupling, though.  I 
would also be careful about using the term "air" here - if it's empty space, 
it's actually vacuum.

> 2) Is the semi isotropic option with 0 compressibility in Z is appropriate
> to keep the pressure fixed? (as shown below in mdp settings)?

That should work.  As long as the dimension along which you need the density 
profile is invariant, you should be able to circumvent any issues.

-Justin

> 3) If the answer to q 2 is negative, whats is appropriate way to shift from
> NPt to NVT run. Do I have to use the equilibrated density from NPT runs and
> edit the last line of gro from -c flag of NPT.gro to obtain the correct
> density? The reason I am asking is that since P fluctuations are huge I am
> not sure if the the density of box from last frame of -c NPT.gro output is
> the correct density...
>
> Please advise on the above queries..Thanks a lot
>
>
>
>
>
> Below is the md.mdp contents:
>
> pbc                 =  xyz
>
>
> integrator          =  md
>
> dt                  =  0.001
>
> nsteps              =  20000000
>
> nstcomm             =  100
>
>
>
> ;           Output control
>
> nstenergy           =  100
>
> nstxout             =  100;1
>
>   nstvout             =  0
>
> nstfout             =  0
>
> nstlog              =  1000
>
> nstxtcout           =  1000
>
> ;           Neighbor searching
>
> nstlist             =  10
>
> ns_type             =  grid
>
> ;           Electrostatics/VdW
>
> coulombtype         =  Shift
>
> vdw-type            =  Shift
>
> rcoulomb-switch     =  1
>
> rvdw-switch         =  1 ;0.9
>
> ;           Cut-offs
>
> rlist               =  1.35
>
> rcoulomb            =  1.1   ;1.1
>
> rvdw                =  1.1
>
>
> Tcoupl              =  v-rescale
>
> tc-grps             =  System
>
> tau_t               =  0.1
>
> ref_t               =  425
>
> *   *
>
> *Pcoupl              =  berendsen*
>
> *Pcoupltype          =  semiisotropic                 *
>
> *tau_p               =  1      0.5                *
>
> *compressibility     =  3.5e-5  0*
>
> *ref_p               =  50   ** *
>
>
>
> gen_vel             =  no
>
> gen_temp            =  500.0
>
> gen_seed            =  173529
>
>
>
> constraints         = none
>
> constraint-algorithm = lincs
>
>
> On 31 March 2013 12:47, Justin Lemkul <jalemkul at vt.edu> wrote:
>
>>
>>
>> On 3/31/13 12:27 PM, Elisabeth wrote:
>>
>>> Thank Justin. I am interested in the density profile at the solvent-air
>>> interface. I wanted to see how density changes with position at different
>>> pressures...I have the equilibrated boxes at several pressures obtained
>>> from NPT but I am not sure whether running g_density on the current
>>> simulations cells provides the profile I am after. Do you have any clue
>>> whether or not the cell has to be extended in one direction to obtain the
>>> density profile?
>>>
>>>
>> If you're trying to produce an air-water interface, then indeed you will
>> need some model for air within the unit cell, but there are several
>> practical problems with that.  The first is that g_density does not deal
>> well with changing box vectors, and the density profiles it produces under
>> an NPT ensemble are incorrect (outstanding bug that needs to be fixed,
>> IIRC).  The second is that if you run a simulation under NPT with void
>> space, the unit cell will just compress down until there is no empty space.
>>   Probably the best solution would be to equilibrate under NPT at the
>> desired conditions, then build a new unit cell and run under NVT.  Using
>> NVT will circumvent both problems, and if the proper pressure, density, etc
>> have already been achieved, then you've got what you're after.
>>
>>
>> -Justin
>>
>> --
>> ==============================**==========
>>
>> Justin A. Lemkul, Ph.D.
>> Research Scientist
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.**vt.edu/Pages/Personal/justin<http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin>
>>
>> ==============================**==========
>> --
>> gmx-users mailing list    gmx-users at gromacs.org
>> http://lists.gromacs.org/**mailman/listinfo/gmx-users<http://lists.gromacs.org/mailman/listinfo/gmx-users>
>> * Please search the archive at http://www.gromacs.org/**
>> Support/Mailing_Lists/Search<http://www.gromacs.org/Support/Mailing_Lists/Search>before posting!
>> * Please don't post (un)subscribe requests to the list. Use the www
>> interface or send it to gmx-users-request at gromacs.org.
>> * Can't post? Read http://www.gromacs.org/**Support/Mailing_Lists<http://www.gromacs.org/Support/Mailing_Lists>
>>

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================



More information about the gromacs.org_gmx-users mailing list